Project Summary/Abstract The human reference genome is part of the foundation of modern human biology, providing a singular coordinate system critical to interpret associations between genotype and phenotype in the emerging field of genomic medicine. However, the current primary reference assembly is largely derived from one individual, and thus is incapable of depicting the full extent of sequence variation observed in the human population. This failure of representation is responsible for reference allele bias in mapping, which is worse for some genetic subpopulations, and occasionally leads to accompanying misinterpretation of gene structure and function. To address this it is imperative that we broaden our reference to include haplotype diversity observed within a cohort of high-quality, phased genome assemblies obtained from distinct and diverse subpopulations. Such a cohort would both capture common variation and haplotype structure, both of which are proving to be crucial context for interpretation. To address this need, we aim to determine a scalable, sequencing/assembly protocol using an optimized collection of new, emerging sequencing technologies that were not readily accessible in the initial phase of this project (e.g.Oxford Nanopore and improved linked Illumina reads strategies). Based upon current projections, we conservatively anticipate being able to reduce the current cost of approximately $30,000 per high quality de novo human genome to $15,000, or less, whilst achieving the same or better quality than the best contemporary assemblies. This advance in throughput and production cost, with an emphasis on maintaining or improving quality, will be critical to increase the scale and scope of genome assembly, with the ultimate goal of applying this work to include a large cohort of diverse individuals in the next reference map.